【摘要】 由于卡尔曼滤波对模型精度有很强的依赖性,因此实际的地磁场测量系统中,宜采用多模型自适应卡尔曼滤波,以提高状态估计的准确度。针对并行子滤波器间的数据融合问题,提出了"距离"意义下的数据融合新算法,应用结果表明算法具有实用性。探讨了通过检验并行子滤波器状态估计一致性的模型误差识别方法,并将其应用于监测地磁场测量系统中船舶磁化参数的变化。更多还原

【Abstract】 In practice,multi-model Kalman filter is suitable for shipboard three-component geomagnetic measurement system because Kalman filtering depends too much on the accuracy of state-space model.A novel data fusion algorithm for the states of the parallel filters is introduced based on the meaning of distance.The application shows the efficiency of the algorithm.A method to identify state-space model errors is discussed by judging the uniformity of the state estimation,which can be used to monitor the changing of ship’s magnetic susceptibility.更多还原